Traditional rapidly-exploring random trees(RRT) algorithms typically tend to explore the environment blindly, which possibly causes the decrease in efficiency. For example, in traditional RRT methods, besides the path from the start point to the goal point, other branch paths unrelated to the result are also generated. In order to take advantage of these branch paths and reduce the blindness of exploration, a robot environment exploration strategy based on the combination of information gain and RRT is proposed. This method estimates the information of the nodes in the unknown environment, selects the nodes with the maximum information gain as the sampling nodes and generates the new nodes with the maximum information gain every time for expansion. This strategy enables the robot to explore the unknown environment autonomously, and also reduces the inherent blindness of the traditional RRT algorithm. The simulation results show that the proposed method can effectively and quickly help the robot explore the unknown environment and realize environmental exploration.